Laboratory Measurement

Introduction

Numerous methods have been developed to quantify the levels of pro- and anti-atherogenic lipoproteins in the blood. Lipoproteins measured by many, but not all, of the assays have been shown to be associated with ASCVD risk. The 2013 American College of Cardiology and American Heart Association (ACC/AHA) cholesterol guideline moved away from titration to specific low-density lipoprotein cholesterol (LDL-C) (or non high-density lipoprotein cholesterol (HDL-C)) targets and toward the use of LDL-C as an approximate indicator of therapeutic efficacy and adherence. LDL-C reductions of ≥50% for high-intensity statins and 30% to <50% for moderate intensity statins were recommended as an indicator of therapeutic response. This approach was largely retained in the current (2018) multi-society (including ACC/AHA) cholesterol guideline. Thus, a consistent method of measurement may be more important than a more accurate measurement of achieved lipid or lipoprotein levels.

Lipoprotein (a) [Lp(a)] and apolipoprotein B [apoB] have been shown to provide additional risk prediction information over that provided by conventional risk factors, which include total cholesterol (or LDL-C) and HDL-C. Additional research will be helpful in this regard.

Standard Method: Calculated LDL-C

LDL, HDL, high-density lipoprotein cholesterol (VLDL) and intermediate-density lipoproteins (IDLs) are more difficult (and more expensive) to measure directly. Instead, laboratories measure the cholesterol content of the various plasma fractions as they are separated along a density gradient (e.g., very low-, low- and high-density) and the total triglyceride levels for all the lipoproteins. Thus, discussion of the clinical management of lipids uses the terminology LDL-C, HDL-C, VLDL-C and triglycerides.

The earliest clinical trials measured only total cholesterol due to technical limitations. Gradient methods were later developed but LDL-C was still difficult to measure directly. Therefore, the Friedewald equation was developed to estimate LDL-C levels. Almost all epidemiologic studies and randomized trials have used calculated LDL-C.

Total cholesterol  =  HDL-C + Triglycerides/5 + LDL-C

The Friedewald equation is not accurate in the nonfasting state, when triglycerides are >400 mg/dL, or in type III hyperlipidemia.

Although direct LDL-C methods have been developed, the Friedewald equation still remains the most inexpensive and common way to measure LDL-C. Commercial laboratories generally use standardized assays based on Centers for Disease Control and Prevention reference methods. These methods have a coefficient of variation of <4% for calculated LDL-C, although this rate may be higher in more general settings.

A more elaborate method of calculating LDL-C using varying denominators for triglycerides that more accurately reflect the triglyceride/VLDL-C ratio was developed using a sample of 1,350,980 clinical lipid profiles but has not been validated in independent populations nor stratified for race/ethnicity, obesity, diabetes, or insulin resistance.

Non–HDL-C

Non–HDL-C reflects all the atherogenic lipoproteins, including LDL-C (Figure 5-1). Interestingly, non–HDL-C levels are similar across the various methods used for direct lipoprotein and apolipoprotein measurement (below), suggesting these methods are simply splitting the same pie differently.

Non–HDL-C is calculated from the standard lipid profile and can be performed on fasting or nonfasting specimens when triglycerides are <500 mg/dL.

Non–HDL-C = Total cholesterol - HDL-C

Non–HDL-C is generally 30 mg/dL higher than LDL-C in patients with LDL-C <150-200 mg/dL. Reductions in non–HDL-C correlate well with the reduction in ASCVD risk from statin and fibrate therapy. Although non–HDL-C was recommended as a second target of the therapy in National Cholesterol Education Program (NCEP) Adult Treatment Panel (ATP) III, subsequent randomized trial evidence from AIM-HIGH found no additional ASCVD risk reduction from further lowering of non–HDL-C with niacin once the LDL-C levels were on average 70 mg/dL. This supports the focus on LDL-C reduction in the 2018 multi-society cholesterol guideline.

Non–HDL-C can be used to monitor response to therapy, especially when triglycerides are 400-500 mg/dL and calculated LDL-C is less accurate. However, many laboratories do not calculate non–HDL and so it is usually easier for the busy clinician to use calculated LDL-C when triglycerides are <400 mg/dL.

Enlarge 

Direct LDL

“Direct” refers to methods that do not require processing of the specimen and have been used when triglycerides are as high as 1000 mg/dL. Many direct methods have been developed and appear to measure varying amounts of other lipoprotein fractions.

In fasting samples, direct LDL levels and calculated LDL-C levels using reference methods have similar associations with ASCVD risk in epidemiologic studies. However, direct LDL-C levels are 5-10 mg/dL lower than mean calculated LDL-C levels. Unlike calculated LDL-C, most direct LDL assays do not include Lp(a).

Lipoprotein Quantification

Because lipoproteins are not measured directly in the Friedewald-based methods, alternate methods for quantitation have been proposed. Lipoprotein density is a function of the triglyceride and cholesterol content, therefore LDL-C and HDL-C may not reflect the actual number of atherogenic lipoprotein particles. In some studies, this has led to discordance between the ASCVD risk associated with LDL-C levels and the risk predicted using methods that quantitate the number of lipoprotein particles.

Lipoprotein particle size was at one time proposed as a measure. Early, small studies found small, dense LDL was associated with greater ASCVD risk. However, later larger studies failed to confirm the association after adjusting for insulin resistance and diabetes. Attention then turned to quantification of lipoprotein particle number instead.

No standardized methods for measuring lipoproteins or their subclasses have been established. Several methods quantitate the number of LDL particles, including nuclear magnetic resonance (NMR), gel gradient electrophoresis, density gradient electrophoresis (vertical auto profile or VAP), microfluidic gel electrophoresis (ion mobility), and a variety of other methods. Each method measures different physiochemical properties which has resulted in widely varying levels of agreement among methods. An Agency for Healthcare Research and Quality (AHRQ) review found that concordance among the different methods ranges from 7% to 94% and discouraged the use of these methods for guiding treatment decisions.

Apolipoprotein B and A Quantification

Another approach is to measure apolipoproteins. Each atherogenic lipoprotein particle (VLDL, IDL, LDL, Lp(a)) contains one apolipoprotein B (apoB) (Figure 5-1 and Figure 2-5). Each HDL particle contains one apolipoprotein A-I (apoA-I). In contrast to the direct lipoprotein quantitation approaches, international standards exist for apoB measurement. The apoB immunoassays have similar accuracy. The nuclear magnetic resonance (NMR) and vertical auto profile (VAP) methods are the most commonly used to measure apoA-I and have more variability. ApoB measured by NMR and VAP is about 15% lower than immunoassays.

There may be discordance between LDL-C and apoB levels and this is associated with an increase in ASCVD risk. However, for ASCVD risk prediction purposes, there is insufficient evidence to determine whether knowledge of apoB adds incrementally to the Pooled Cohort Equations or other risk prediction equations to yield a more accurate estimate of ASCVD risk to guide therapeutic decisions. Although apoB levels correlate well with ASCVD risk reduction from statin therapy, the incremental additional risk reduction from apoB is similar to that for non–HDL-C. Taken together, these findings suggest there is no particular benefit for measuring apoB in the routine management of cholesterol to reduce ASCVD risk. When apoB is measured, a level of ≥130 mg/dL is considered equivalent to an LDL-C level of ≥160 mg/dL, and is therefore considered a risk enhancing factor according to the 2018 multi-society cholesterol guideline.

ApoB levels may be helpful for diagnosis of hereditary lipid disorders (see Familial Hyperlipidemias). In patients with severe hypertriglyceridemia due to a chylomicron disorder, apoB levels will be low to average; in those with familial combined hyperlipidemia or type III hyperlipidemia, apoB levels will be elevated due to increased levels of VLDL and IDL.

Enlarge 

Enlarge 

Lp(a)

Lipoprotein (a) (Lp(a)) measurement has been problematic due to interindividual variation in the number of kringle domains in the apo(a) moiety. Methods have been standardized that are apo(a) insensitive. However, there are large race-dependent variations in the association between Lp(a) levels and ASCVD risk. Reference ranges for population groups other than those of European ancestry are only now being developed (see Risk Assessment: Primary Prevention LDL-C <190 mg/dL). The 2018 multi-society cholesterol guideline states that an elevated (≥50 mg/dL) Lp(a) level should be considered a risk enhancer.

Clinical Highlights

• LDL-C
  • Monitor response and adherence to therapy with calculated LDL-C from a standard method when triglycerides are <400 mg/dL. It is inexpensive and readily available.
• Triglycerides
  • Triglycerides >500 mg/dL
  • Triglycerides 400-500 mg/dL
    • Encourage the patient to improve lifestyle and retest.
    • Use non–HDL-C to monitor ASCVD risk reduction therapy.
    • Direct LDL-C is also an option, using the same assay each time and acknowledging the issues regarding accuracy.
  • Triglycerides 150-400 mg/dL: use LDL-C to monitor therapy.
• ApoB
  • When apoB is measured, a level of ≥130 mg/dL is considered a risk enhancing factor and should be included in the shared decision-making on initiating LDL-C-lowering therapy.
• Lp(a)
  • If measured, elevated Lp(a) levels (≥50 mg/dL) from a standardized assay may be an additional factor to consider in the clinician-patient discussion when deciding to initiate statin or other therapies.